Open access
Aug 2026
SETAS-VAD: Semantically Enriched Text-Aligned Scoring for Weakly Supervised Video Anomaly Detection
Under fully reproducible conditions on UCF-Crime and XD-Violence, SETAS-VAD achieves state-of-the-art temporal localization, with per-threshold gains increasing at stricter IoU values, indicating improved boundary precision rather than coarse detection sensitivity.
Mohamed Mahmoud, Mostafa Farouk Senussi, Mahmoud Abdalla et al.
· Mathematics · 0 citations